Auditable reproduction of Qwen2.5-7B vs SimpleRL-Zoo on MATH500 for Limit of RLVR (arXiv:2504.13837 v5).
-
Updated
Sep 1, 2026 - Python
#
math500
Here are 2 public repositories matching this topic...
A small experiment repository comparing a base reasoning model against RLVR-GRPO checkpoints on the Math500 dataset. It includes evaluation results, short-form observations, and a local temp_clone of the full open-posttraining-system codebase for reference.
reinforcement-learning post-training evaluating-models policy-optimization sparse-rewards reasoning-models rlvr-grpo math500 grpo-checkpoint open-posttraining-system
-
Updated
Jun 17, 2026 - Jupyter Notebook
Add this topic to your repo
To associate your repository with the math500 topic, visit your repo's landing page and select "manage topics."